Despite calls for universal colonoscopy (CY) for colorectal cancer (CRC) screening, its clinical application will be limited by issues of cost, safety, constrained resources, and patient acceptance. A method to quantify risk for CRC and advanced, pre-malignant polyps (collectively known as "advanced neoplasia") could allow use of different screening tests (CY, sigmoidoscopy, fecal occult blood) to be tailored to risk. While epidemiological studies had demonstrated associations between several phenotypic and environmental factors and the risks for CRC and polyps, whether these factors can be used in combination to estimate and stratify individual patient risk has not been determined. The objective of this research is to create a system to quantify risk for advanced neoplasia (AN) in average risk persons; the long-term goal is to improve the efficiency and cost-effectiveness of CRC screening. Using the clinical sites for two employer-based programs of screening CY, one of which has already produced important research about CRC screening, we propose to establish a research infrastructure and blood (serum and DNA) repository. Building on our preliminary work, we will collect detailed and reliable data prospectively from 6,000 average-risk persons undergoing first-time screening CY. One week prior to CY, patients will receive a survey in the mail to measure demographic, clinical, familial, and lifestyle (diet, exercise, medication use, etc.) factors. Research assistants may assist patients with the survey by telephone. Physical measures and blood specimen donation will be obtained by trained clinical personnel immediately prior to CY. Endoscopic and histological data will be collected following CY. Using methods of multivariable analysis to identify features that predict risk for AN, we will derive two clinical prediction rules (CPRs) to stratify risk for AN 1) anywhere in the large intestine--to target the timing and intensity of initial screening, and; 2) in the proximal colon--to target use of CY after sigmoidoscopy. The hypothesis for this research, supported by our preliminary data, is that risk factors from epidemiological studies may be used to create CPRs that quantify risk for AN and thus, to tailor CRC screening. This project will 1) be the largest study of community-based screening CY in men and women; 2) measure comprehensively and test simultaneously the predictive value of multiple risk factors for colorectal neoplasia; 3) establish CPRs based on these factors, and 4) establish a serum and DNA repository for future analysis of serum markers and genetic profiles that may refine the ability to stratify individual risk. Attaining these objectives will allow tailoring of initial CRC screening according to risk, with targeted use of CY among higher-risk persons, and reduction of unnecessary CY for very low-risk persons. The knowledge generated by this research will help maximize prevention and minimize the morbidity and cost of CRC screening.